I. Aims: We have previosly shown that human lymphocytes can be sensitized in vitro to recognize HLA identical leukemia cells. These antileukemic effector cells can destroy leukemic blasts in vitro, yet do not kill remission lymphocytes from the leukemic patient. We now propose to further characterize and manipulate these in vitro antileukemic immune responses. II. Appoach: Leukemic blasts will be obtained from patients by bone marrow aspiration at diagnosis and at relapse. Cryopreserved blasts will be used as stimulating and target cells in mixed lymphocyte cultures and cell mediated lysis assays. Responding cells in these in vitro cultures will be autologous remission lymphocytes or lymphocytes from HLA identical siblings. To facilitate studying antileukemic responses, we address methodologic problems using a model system in which helper and cytotoxic cellular responses are stimulated by autologous lymphocytes labelled with bacterial antigen. III. Objectives: a) To examine the induction requirements, antigenic specificity, and immuno-regulation of human lymphocytes primed in vitro to destroy autologous and histocompatible leukemic blasts. b) To evaluate relationships between this in vitro response and the clinical course of leukemic patients. c) To use newer culture methods to generate large quantities of polyclonal, and single cell derived-monoclonal populations of antileukemic lymphocytes; and examine their specificity in suppressive, cytotoxic, and helper systems. d) To identify the immunologic properties and differentiation markers of human antileukemic effector lymphocytes obtained from bulk and cloned cultures. e) To generate adequate quantities of these antileukemic cells to initiate clinical trials with specific adoptive immunotherapy in selected high risk leukemic patients.